Title: Design of intelligent system for indoor illumination adjustment based on deep learning

Authors: Chen Qun Wu

Addresses: College of Art and Design, Yellow River Conservancy Technical Institute, Kaifeng 475000, China

Abstract: In order to overcome the low adjustment accuracy and efficiency of the traditional regulation system, this paper designed an indoor lighting intensity intelligent regulation system based on deep learning. The hardware part of the system is designed by deep learning. Then, based on the analysis of sensor data and historical data, the corresponding intelligent adjustment table is formed. After the convolution and pooling operation, the training samples are combined with restricted Boltzmann machine. At the same time, the natural illumination model is built based on the time cycle variation characteristics of sunlight, and the indoor and outdoor illumination is calculated with the deep learning results, so as to obtain the brightness level of dimming and to realise intelligent regulation. The experimental results show that the intelligent adjustment accuracy of the system is between 95.0% and 98.5%, and the adjustment efficiency is always above 95%.

Keywords: deep learning; indoor illumination; illumination adjustment; illumination model.

DOI: 10.1504/IJISE.2023.128676

International Journal of Industrial and Systems Engineering, 2023 Vol.43 No.2, pp.137 - 152

Received: 26 Oct 2020
Accepted: 08 Mar 2021
Published online: 01 Feb 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article